Jouni Malinen <jouni@xxxxxxxxxxxxxx> wrote: > When doing "rmmod ath11k_pci", ath11k performs global SOC reset > and MHI reset, where 0 address access is captured by IOMMU. See > log below: > > ... > [ 133.953860] ath11k_pci 0000:02:00.0: setting mhi state: DEINIT(1) > [ 133.959714] ath11k_pci 0000:02:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x000a address=0x0 flags=0x0020] > [ 133.973854] ath11k_pci 0000:02:00.0: MHISTATUS 0xff04 > [ 133.974095] ath11k_pci 0000:02:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x000a address=0x0 flags=0x0020] > ... > > This issue is also observed in SSR process, cause a similar > sequence as above is performed. > > Such an invalid access occurs because, during rmmod or SSR, MSI > address is cleared but HW MSI functionality not disabled, thus HW > target is able to raise an MSI transaction with 0 as MSI address. > > So it can be fixed by simply disabling MSI before reset. For SSR, > since MSI functionality is still needed after target is brought > back, we need to reenable it. > > Also change naming of some interfaces related. > > Tested-on: QCA6390 hw2.0 PCI WLAN.HST.1.0.1-01740-QCAHSTSWPLZ_V2_TO_X86-1 > Tested-on: WCN6855 hw2.0 PCI WLAN.HSP.1.1-01720.1-QCAHSPSWPL_V1_V2_SILICONZ_LITE-1 > > Signed-off-by: Baochen Qiang <bqiang@xxxxxxxxxxxxxx> > Signed-off-by: Jouni Malinen <jouni@xxxxxxxxxxxxxx> > Signed-off-by: Kalle Valo <kvalo@xxxxxxxxxxxxxx> Patch applied to ath-next branch of ath.git, thanks. 96527d527b27 ath11k: Handle MSI enablement during rmmod and SSR -- https://patchwork.kernel.org/project/linux-wireless/patch/20210913180246.193388-5-jouni@xxxxxxxxxxxxxx/ https://wireless.wiki.kernel.org/en/developers/documentation/submittingpatches
